Bridging Project 4: Haloacid Dehalogenase (HAD) Superfamily

HAD超家族是一个大的酶家族（约19000个非冗余序列），有100多个磷酸转移酶（磷酸脲酶、atp酶和磷酸酶）存在于所有三个生命体中[2-4]，并且在每个细胞中都有大量同源物（大肠杆菌28个，鼠伤寒沙门氏菌35个，铜绿假单胞菌31个，结核分枝杆菌30个，蜡样芽孢杆菌31个，脆弱杆菌24个，大肠杆菌35个）。

The HAD superfamily is a large enzyme family (~19,000 nonredundant sequences) [1] of phosphotransferases (phosphomutases, ATPases and phosphatases) represented in all three kingdoms of life [2-4], and, within each cell, by a large number of homologs (28 in E. coli; 35 in Salmonella typhimurium; 31 in Pseudomonas aeruginosa; 30 in Mycobacterium tuberculosis; 31 in Bacillus cereus; 24 in Bacteroides fragilis; 24 in Streptococcus pneumoniae; 45 in Saccharomyces cerevisiae; 84 in Caenorhabditis elegans; 169 in Arabidopsis thaliana; 292 in Selaginella moeltendorffii; 183 in human). As many as 80-90% of the members are phosphatases [5], the vast majority of which have unknown functions. Approximately 40% of the bacterial metabolome is comprised of phosphorylated metabolites [6]. Phosphate substituents are common because they enhance the water solubility of the metabolite as well as its ability to bind to metabolic enzymes with high affinity and specificity. The removal of phosphate groups from phosphorylated metabolites is performed by phosphatases. The "function" of a particular phosphatase is defined by the phosphorylated metabolite that it targets in the cell, i.e., by its "physiological substrate". Thus, the HAD phosphatases meet the demands of cellular processes and metabolic pathways that involve phosphorylated macromolecules and metabolites. Divergence in HAD phosphatase funcfion is based on the divergence of the substrate-recognition elements. The substrate-recognition elements are separate from the catalytic scaffold, which is located in the core domain (Figure 1A). The four pepfide segments or "motifs" which form the active site position the consen/ed Asp nucleophile, Asp acid/base, the Lys/Arg and Ser/Thr phosphate-binding residues and the Mg^* cofactor Asp/Glu binding residues (Figure IB). These residues, in combinafion with the scaffold main-chain elements, form a steric and electrostafic mold that stabilizes the trigonal bipyramidal transifion states/intermediates produced along the reaction pathway (Figure 1B) [7]. The HAD phosphatase substrate recognition elements are located in either a cap domain (as in HAD classes Cl and C2, also known as Type I and Type 11) tethered to the core domain by a solvated linker, or in short loop/helical segments that extend from the core domain (as in the "capless" HAD class CO also known as Type III) (Figure 1A) [8]. Although HAD phosphatases possess the same catalytic site and proceed through the same second partial reaction, they are able to use the specific structural requirements of the substrate-binding step and the subsequent addition-eliminafion steps of the first partial reacfion to discriminate between the physiological substrate and other phosphorylated species (macromolecules and metabolites). The induced fit model, wherein substrate binding is followed by cap domain or loop closure, applies to most HAD phosphatases. Favorable electrostafic interaction between the substrate leaving group and the cap domain/gafing loops will contribute to the substrate-binding affinity. For efficient turnover, the phosphoryl group must be bound in the correct orientation within the catalyfic site. If the substrate-leaving group is too large or too small, nonproductive binding is likely to occur. Thus, the size, shape and electrostafic surface ofthe acfive site region that extends from the catalytic site to the active site entrance can provide significant insight into the identity of the physiological substrate. This serves as the basis for the use of virtual screening (made possible by the Structure Core and Computation Core) to identify candidates for the physiological substrate herein. Substrate specificities defined by experimental activity screens suggest that the typical HAD phosphatase has loose substrate specificity coupled with modest catalytic efficiency. Thus, acfivity screens alone often cannot idenfify the actual physiological substrate. Rather, they provide candidates that can be further interrogated using the tools provided by the Sequence/Genome Analysis Core and Microbiology Core.